Singleton Method – Python Design Patterns
Prerequisite: Singleton Design pattern | Introduction
What is Singleton Method in Python
Singleton Method is a type of Creational Design pattern and is one of the simplest design patterns available to us. It is a way to provide one and only one object of a particular type. It involves only one class to create methods and specify the objects.
Singleton Design Pattern can be understood by a very simple example of Database connectivity. When each object creates a unique Database Connection to the Database, it will highly affect the cost and expenses of the project. So, it is always better to make a single connection rather than making extra irrelevant connections which can be easily done by Singleton Design Pattern.
singleton-pattern
Definition: The singleton pattern is a design pattern that restricts the instantiation of a class to one object.
Now let’s have a look at the different implementations of the Singleton Design pattern.
Method 1: Monostate/Borg Singleton Design pattern
Singleton behavior can be implemented by Borg’s pattern but instead of having only one instance of the class, there are multiple instances that share the same state. Here we don’t focus on the sharing of the instance identity instead we focus on the sharing state.
# Singleton Borg pattern
class Borg:
# state shared by each instance
__shared_state = dict()
# constructor method
def __init__(self):
self.__dict__ = self.__shared_state
self.state = 'GeeksforGeeks'
def __str__(self):
return self.state
# main method
if __name__ == "__main__":
person1 = Borg() # object of class Borg
person2 = Borg() # object of class Borg
person3 = Borg() # object of class Borg
person1.state = 'DataStructures' # person1 changed the state
person2.state = 'Algorithms' # person2 changed the state
print(person1) # output --> Algorithms
print(person2) # output --> Algorithms
person3.state = 'Geeks' # person3 changed the
# the shared state
print(person1) # output --> Geeks
print(person2) # output --> Geeks
print(person3) # output --> Geeks
Output:
Algorithms
Algorithms
Geeks
Geeks
Geeks
Singleton-Design-pattern
Double Checked Locking Singleton Design pattern
It is easy to notice that once an object is created, the synchronization of the threading is no longer useful because now the object will never be equal to None and any sequence of operations will lead to consistent results.
So, when the object will be equal to None, then only we will acquire the Lock on the getInstance method.
# Double Checked Locking singleton pattern
import threading
class SingletonDoubleChecked(object):
# resources shared by each and every
# instance
__singleton_lock = threading.Lock()
__singleton_instance = None
# define the classmethod
@classmethod
def instance(cls):
# check for the singleton instance
if not cls.__singleton_instance:
with cls.__singleton_lock:
if not cls.__singleton_instance:
cls.__singleton_instance = cls()
# return the singleton instance
return cls.__singleton_instance
# main method
if __name__ == '__main__':
# create class X
class X(SingletonDoubleChecked):
pass
# create class Y
class Y(SingletonDoubleChecked):
pass
A1, A2 = X.instance(), X.instance()
B1, B2 = Y.instance(), Y.instance()
assert A1 is not B1
assert A1 is A2
assert B1 is B2
print('A1 : ', A1)
print('A2 : ', A2)
print('B1 : ', B1)
print('B2 : ', B2)
Output:
A1 : __main__.X object at 0x02EA2590
A2 : __main__.X object at 0x02EA2590
B1 : __main__.Y object at 0x02EA25B0
B2 : __main__.Y object at 0x02EA25B0
Creating a singleton in Python
In the classic implementation of the Singleton Design pattern, we simply use the static method for creating the getInstance method which has the ability to return the shared resource. We also make use of the so-called Virtual private Constructor to raise the exception against it although it is not much required.
# classic implementation of Singleton Design pattern
class Singleton:
__shared_instance = 'GeeksforGeeks'
@staticmethod
def getInstance():
"""Static Access Method"""
if Singleton.__shared_instance == 'GeeksforGeeks':
Singleton()
return Singleton.__shared_instance
def __init__(self):
"""virtual private constructor"""
if Singleton.__shared_instance != 'GeeksforGeeks':
raise Exception("This class is a singleton class !")
else:
Singleton.__shared_instance = self
# main method
if __name__ == "__main__":
# create object of Singleton Class
obj = Singleton()
print(obj)
# pick the instance of the class
obj = Singleton.getInstance()
print(obj)
Output:
__main__.Singleton object at 0x014FFE90
__main__.Singleton object at 0x014FFE90
Class diagram
Class Diagram of Singleton Design Pattern
singleton-class-diagram
Advantages of using the Singleton Method:
- Initializations: An object created by the Singleton method is initialized only when it is requested for the first time.
- Access to the object: We got global access to the instance of the object.
- Count of instances: In singleton, method classes can’t have more than one instance
Disadvantages of using the Singleton Method:
- Multithread Environment: It’s not easy to use the singleton method in a multithread environment, because we have to take care that the multithread wouldn’t create a singleton object several times.
- Single responsibility principle: As the Singleton method is solving two problems at a single time, it doesn’t follow the single responsibility principle.
- Unit testing process: As they introduce the global state to the application, it makes the unit testing very hard.
Applicability
- Controlling over global variables: In the projects where we specifically need strong control over the global variables, it is highly recommended to use Singleton Method
- Daily Developers use: Singleton patterns are generally used in providing the logging, caching, thread pools, and configuration settings and are often used in conjunction with Factory design patterns.
Further read: Singleton method in Java, Singleton Design Pattern Practices with Examples
